Improving stroke diagnosis accuracy using hyperparameter optimized deep learning

• Main Authors: Tessy Badriyah, Dimas Bagus Santoso, Iwan Syarif, Daisy Rahmania Syarif
• Format: Article
• Language: English
• Published: Universitas Ahmad Dahlan 2019-11-01
• Series: IJAIN (International Journal of Advances in Intelligent Informatics)
• Subjects: feature selection; deep learning; hyperparameter optimization
• Online Access: http://ijain.org/index.php/IJAIN/article/view/427
• ISSN: 2442-6571; 2548-3161

Stroke may cause death for anyone, including youngsters. One of the early stroke detection techniques is a Computerized Tomography (CT) scan. This research aimed to optimize hyperparameter in Deep Learning, Random Search and Bayesian Optimization for determining the right hyperparameter. The CT scan images were processed by scaling, grayscale, smoothing, thresholding, and morphological operation. Then, the images feature was extracted by the Gray Level Co-occurrence Matrix (GLCM). This research was performed a feature selection to select relevant features for reducing computing expenses, while deep learning based on hyperparameter setting was used to the data classification process. The experiment results showed that the Random Search had the best accuracy, while Bayesian Optimization excelled in optimization time.